-
公开(公告)号：EP4181559A1
公开(公告)日：2023-05-17
申请号：EP21842932.2
申请日：2021-06-17
发明人： WANG, Jian , XU, Chen , YU, Tianhang , LI, Rong , WANG, Jun
IPC分类号： H04W24/02
摘要： This application provides a communication method and apparatus. The method may include: A first communication apparatus receives configuration information of a first neural network; the first communication apparatus determines the first neural network based on the configuration information of the first neural network; the first communication apparatus obtains first channel information based on channel information obtained through measurement and the first neural network, where a data amount of the first channel information is less than a data amount of the channel information obtained through measurement; and the first communication apparatus sends the first channel information, where the first channel information is used to obtain second channel information by using a second neural network, and the second channel information is used for data transmission. According to the method provided in embodiments of this application, channel information can be fed back, and a system throughput can be improved.
-
公开(公告)号：EP4120042A1
公开(公告)日：2023-01-18
申请号：EP21788348.7
申请日：2021-04-16
发明人： ZHANG, Gongzheng , HU, Bin , XU, Chen , WANG, Jian , LI, Rong
IPC分类号： G05D1/02
摘要： A decision-making method for an agent action and a related device are provided, and are used in the field of communication technologies. The method includes: A first agent processes first state information obtained from an environment through a first model, to obtain a first cooperation message (501); the first agent sends the first cooperation message (502) to at least one second agent; the first agent receives second cooperation message (503) sent by the at least one second agent; the first agent processes the first cooperation message and the second cooperation message through a second model, to obtain a first cooperation action (504) performed by the first agent, where the second cooperation message is sent by the at least one second agent. In this way, an original parallel network model that obtains both a cooperation message and a cooperation action based on state information becomes a serial network model, a structure of each network model is simpler, and a neural network model is trained more easily.
-
公开(公告)号：EP3637731A1
公开(公告)日：2020-04-15
申请号：EP18823331.6
申请日：2018-06-26
发明人： HUANG, Lingchen , LI, Rong , ZHANG, Huazi , XU, Chen , DAI, Shengchen , ZHANG, Gongzheng , QIAO, Yunfei
IPC分类号： H04L29/08
摘要： A polar code encoding method and apparatus are disclosed to provide a new check encoding scheme. The method includes: separately performing check encoding on the at least two to-be-checked first bit sequences to obtain at least two check bit sequences, where a union set of the at least two to-be-checked first bit sequences includes the K information bits, and K is a positive integer; interleaving the K information bits and the at least two check bit sequences, or interleaving a first part of information bit sequence and the first check bit sequence to obtain an interleaved third bit sequence, where the second check bit sequence, the third check bit sequence, and a second part of information bit sequence in the sequence of all information bits except the first part of information bit sequence form a second bit sequence; and performing polar encoding on the second bit sequence.
-
公开(公告)号：EP4358446A1
公开(公告)日：2024-04-24
申请号：EP22836925.2
申请日：2022-07-05
发明人： HU, Bin , WANG, Jian , XU, Chen , ZHANG, Gongzheng , LI, Rong
摘要： This application provides a model training method and a related apparatus, to help improve a convergence speed of model training and improve end-to-end communication quality. The method includes: The first communication apparatus sends first data to a second communication apparatus through a channel, where the first data is an output result of the first machine learning model. The second communication apparatus receives second data through a channel, and inputs the second data into a second machine learning model, to obtain third data; determines a first loss function based on the third data and the first training data; and sends the first loss function to the first communication apparatus through a feedback channel. The first communication apparatus receives a second loss function through a feedback channel, and updates a parameter of a control layer based on Kalman filtering and according to the second loss function, to obtain an updated parameter of the control layer, where the updated parameter of the control layer is used to update a parameter of the first machine learning model.
-
公开(公告)号：EP3934138A1
公开(公告)日：2022-01-05
申请号：EP21177059.9
申请日：2018-03-24
发明人： XU, Chen , LI, Rong , ZHANG, Gongzheng , ZHOU, Yue , HUANG, Lingchen
摘要： Embodiments of this application provide a polar coding method and a coding apparatus. The method includes: obtaining a to-be-sent information block and a target code length M of a polar code; determining a mother code length N used for polar encoding, when the target code length M is greater than N, if an encoding parameter of the information block meets a preset condition, segmenting a to-be-encoded information bit sequence into p subsegments, and performing independent polar encoding on the p subsegments, to obtain p encoded bit sequences whose lengths are respectively mother code lengths of the subsegments; separately performing rate matching on p encoding results, to obtain p encoded bit sequences whose lengths are respectively target code lengths of the subsegments; and combining the p rate-matched encoded bit sequences, to obtain an encoded bit sequence whose length is M, where p is an integer greater than or equal to 2. The coding method can reduce use times of a repetition-based rate matching scheme, and reduce a performance loss caused by repetition.
-
公开(公告)号：EP3930278A1
公开(公告)日：2021-12-29
申请号：EP20773203.3
申请日：2020-02-28
发明人： WANG, Jian , XU, Chen , LI, Rong , QIN, Dali , WANG, Jun
IPC分类号： H04L29/02
摘要： According to a scheduling method and apparatus in a communication system, and a storage medium that are provided in embodiments of this application, a communication device obtains system status information, where the system status information includes network status information; obtains a scheduling policy based on the system status information and a deep neural network; and performs communication according to the scheduling policy. The deep neural network is obtained through training based on historical system status information, and the historical system status information includes system status information in all scheduling periods before a current scheduling period. Therefore, the scheduling policy obtained based on the deep neural network can meet a balancing requirement of throughput and fairness and solves a problem of low performance of an existing communication system.
-
公开(公告)号：EP3879709A1
公开(公告)日：2021-09-15
申请号：EP19890969.9
申请日：2019-11-26
发明人： XU, Chen , LI, Rong , YU, Tianhang , QIAO, Yunfei , DU, Yinggang , HUANG, Lingchen , WANG, Jun
摘要： Embodiments of this application provide an encoding method and apparatus, and a decoding method and apparatus, and relate to the communications field. In the method, a corresponding neural network unit may be generated based on a kernel matrix, and then an encoding neural network or a decoding neural network may consist of neural network units. In this way, the encoding neural network or the decoding neural network is obtained after small neural network units are connected, so that in an encoding/a decoding learning process, generalization can be implemented to entire codeword space by using a small learning sample, to weaken impact of information with a relatively long codeword on complexity and learning difficulty of a neural network.
-
公开(公告)号：EP3683990A1
公开(公告)日：2020-07-22
申请号：EP18873004.8
申请日：2018-09-25
发明人： ZHAO, Mingmin , ZHANG, Gongzheng , XU, Chen , LI, Rong
摘要： This application provides a data retransmission method and apparatus. The method includes: A transmitting device obtains information to be transmitted for a t th time, where the information to be transmitted for the t th time includes R t extension locations and information to be transmitted for a (t-1) th time, and the extension locations include M t information bits and L t check bits corresponding to the M t information bits. The transmitting device then performs Polar encoding on the information to be transmitted for the t th time, to obtain a codeword after the Polar encoding, obtains a codeword for (t-1) th retransmission based on the codeword after the Polar encoding, and transmits the codeword for (t-1) th retransmission. A receiving device performs polar decoding after receiving the codeword for (t-1) th retransmission, to obtain a decoding result of codewords for t times of transmission. By performing, on an encoding side, check encoding on the information bits in an extension part, a decoding path can be reduced in a decoding process, thereby greatly reducing decoding complexity, and reducing storage overheads and calculation overheads.
-
公开(公告)号：EP3614568A1
公开(公告)日：2020-02-26
申请号：EP18794746.0
申请日：2018-05-03
发明人： HUANG, Lingchen , ZHANG, Huazi , LI, Rong , ZHANG, Gongzheng , XU, Chen
摘要： An encoding method and apparatus are provided, to provide a new check encoding manner. The method includes: performing, by a transmit end, check encoding on to-be-encoded information to obtain a check encoding codeword, where in the check encoding codeword, a length of an information bit is K, and a length of a check bit is J; performing, by the transmit end, an interleaving operation on the check encoding codeword, where an interleaving sequence S used in the interleaving operation includes J subsequences, and an i th subsequence includes a position index of an element 1 in an intermediate result vector Ti and a value of (K+i), where 1≤i≤J, i is an integer, T i =(∼M)&(V i ), M=M|(V i ), V i is a column vector of a checking part matrix P, P is a submatrix of a system-form generator matrix G for check encoding, ∼ represents a bit-by-bit NOT operation, & represents a bit-by-bit AND operation, and | represents a bit-by-bit OR operation; and performing, by the transmit end, polar encoding on a check encoding codeword obtained after the interleaving operation.
-
公开(公告)号：EP3598673A1
公开(公告)日：2020-01-22
申请号：EP18770517.3
申请日：2018-03-24
发明人： XU, Chen , LI, Rong , ZHANG, Gongzheng , ZHOU, Yue , HUANG, Lingchen
摘要： Embodiments of this application provide a polar coding method and a coding apparatus. The method includes: obtaining a to-be-sent information block and a target code length M of a polar code; determining a mother code length N used for polar encoding, when the target code length M is greater than N, if an encoding parameter of the information block meets a preset condition, segmenting a to-be-encoded information bit sequence into p subsegments, and performing independent polar encoding on the p subsegments, to obtain p encoded bit sequences whose lengths are respectively mother code lengths of the subsegments; separately performing rate matching on p encoding results, to obtain p encoded bit sequences whose lengths are respectively target code lengths of the subsegments; and combining the p rate-matched encoded bit sequences, to obtain an encoded bit sequence whose length is M, where p is an integer greater than or equal to 2. The coding method can reduce use times of a repetition-based rate matching scheme, and reduce a performance loss caused by repetition.
-
-
-
-
-
-
-
-
-
搜索结果
36 条记录 (耗时 0.022 秒)
